Research On Global Path Planning Algorithm Of Mobile Robot In Complex And Dynamic Environment

Path planning provides the information of the Lunar Rover to the motion control during moving,which is one of the key elements of studying and developing of the whole Lunar Rover system.Lunar Rover is also a special mobile robot working in the complex lunar environment. Therefore,Researches on the path planning of mobile robot in complex and dynamic environment are very important to the study of the path planning of Lunar Rover.With the background that project phaseⅡand phaseⅢof lunar exploration's demonstration developed by Commission of Science Technology and Industry for National Defense,this paper researches the mobile robot's traditional global path planning algorithms, global path planning algorithms in the complex and dynamic environment, simulates lunar rover's path planning processing,and the cost evaluation of path planning.The main content of this paper is listed as the followings:At first,the two traditional mobile robot's global path planning algorithm A~* and D~* Lite were carried out.Aimed at engineering constraints,this paper introduced the steering angle and turning frequency into the evaluation function and brought up a A~* algorithm-based global path planning algorithm A~* angle which made the planning path smoother, more effective.At last,the new algorithm was simulated on the MATLAB simulation platform.Secondly,aimed at some problems caused by the traditional algorithms when they were used in the three-dimensional complex and dynamic environment,this paper proposed a new method of three-dimensional modeling and directional three-dimensional terrain judging methods,studied the industrial principles and prediction rolling principle and amended the traditional global path planning algorithm D~* Lite.So that it could be used in the three-dimensional complex and dynamic environment.Thirdly,aimed at the "semi-autonomous and teleoperation" working mode of our country's Lunar Rover,this paper carried out the typical local obstacle avoidance algorithm Morphin and combined with global path planning algorithm D~* Lite to simulate Lunar Rover's path planning processing on the MATLAB simulation platform.Lastly,based on the modification of the evaluation function,this paper presented a comprehensive evaluation method of the paths instead of the traditional criterion which only took the path length into account. This new criterion considered the cost of mobile robot to walk straight, the cost of turning and adjustment.These costs will be integrated for power consumption and time-consuming and provided a standard for the performance of the paths.